Skip to main content

Two Week Case Analyses Explainer

Starting in Week 5, Case Studies stretch across two weeks instead of one. Your group picks a broad domain (below), then researches and settles on one specific, documented case within it: a named tool, deployed by a named organization, used in a specific way. The domain lists below are meant to get you oriented and thinking; they are not a menu of pre-picked cases. Finding the specific case is part of the assignment.

There are three two-week rounds. Each round offers a handful of large domains to choose from; your group picks one domain and one case inside it.

Weeks 5–6 (Full Case I)#

  • Healthcare — diagnostic and medical-imaging tools; clinical documentation and “ambient scribe” systems; insurance claims and prior-authorization algorithms; mental-health chatbots and triage tools; hospital staffing and resource-allocation systems.
  • Art & Creative Industries — generative image and video tools; AI in film/animation production pipelines; music generation and voice cloning; authorship and copyright disputes; museum and curatorial applications.
  • Education — proctoring and academic-integrity software; admissions and enrollment screening; adaptive tutoring and personalized-learning platforms; automated grading and feedback tools; AI-detection tools used against students.
  • Athletics & Sports — performance and biometric tracking; officiating and instant-replay systems; scouting and recruiting analytics; fan engagement and broadcast AI; injury-prediction systems.
  • Politics & Government — elections, campaign content, and political deepfakes; public-input and civic-engagement tools; generative AI used inside agencies; benefits eligibility and fraud-detection systems; infrastructure and public-service planning.

Weeks 7–8 (Full Case II)#

  • Law & Criminal Justice — predictive policing; risk-assessment and sentencing tools; e-discovery and legal-research assistants; facial recognition used in investigations; parole and prison-management systems.
  • Finance & Banking — credit scoring and loan underwriting; fraud detection; algorithmic trading; robo-advisors; insurance underwriting and claims processing.
  • Journalism & Media — AI-generated news and summarization; content moderation; deepfake detection; recommendation feeds that shape what gets read; automated fact-checking.
  • Hiring & Employment — resume screening and applicant-tracking systems; video-interview analysis; gig-work scheduling and dispatch algorithms; workplace productivity monitoring; performance-review or termination systems.
  • Military & Defense — autonomous weapons and targeting systems; surveillance and intelligence analysis; logistics and supply-chain optimization; decision-support tools for command; military cybersecurity.
  • Scientific Research — protein-structure and materials prediction; AI-assisted data analysis and discovery; automated and “self-driving” laboratories; AI-written papers, fabricated citations, and peer review; research-integrity and paper-mill detection tools.

Weeks 9–10 (Full Case III)#

  • Agriculture — crop and yield prediction; autonomous farm equipment; precision irrigation and pesticide application; livestock monitoring; labor and worker tracking on farms.
  • Transportation — autonomous/self-driving vehicles; traffic and logistics optimization; ride-share dispatch and pricing; public-transit planning tools; aviation and air-traffic systems.
  • Retail & Consumer — recommendation engines; dynamic and surge pricing; inventory and supply-chain forecasting; self-checkout and loss-prevention surveillance; customer-service chatbots.
  • Entertainment & Gaming — content-recommendation feeds; generative game content and NPC behavior; matchmaking algorithms; synthetic/deepfake performers; streaming-platform moderation.
  • Environmental & Climate Science — climate modeling and forecasting; disaster prediction and early-warning systems; resource allocation for water and energy grids; conservation and wildlife monitoring; environmental-compliance and emissions monitoring.
  • An undiscussed area of your choosing — any domain not covered in Full Cases I or II. Your case must meet the same standard: a named tool, a named organization, and a documented deployment. Name your domain and case in the Week 9 Update so the instructional team can confirm it.

Turning In#

Each two-week round has two deliverables, not one big one at the end:

  1. End of Week 1 — Update (Sunday): Your group turns in an update answering a short set of scaffolding questions — e.g., what specific case you’ve settled on, what sources you’ve found so far, and which DTPA areas you can already speak to versus still need to research. This checkpoint exists to catch groups that are stuck on a domain instead of a case, or missing sources, while there’s still a week left to fix it.
  2. End of Week 2 — Full Case (Sunday): The complete Case Study, using the same template and DTPA framework as the weekly analyses, is due the Sunday after the Update.

Check the Schedule for the exact dates of the Update and Full Case deadlines for each round.

Choosing Your Case#

  • Pick a domain your group actually wants to spend two weeks in. You’ll be with it (and the same groupmates) for the full round.
  • “Domain” is not “case.” “AI in healthcare” is not a case; a specific hospital system’s use of an ambient-documentation tool is. Narrow down quickly so you have time left to research, not just to search.
  • Use the same sourcing standard as the weekly Case Studies: named organization, named tool where possible, documented deployment not a general trend piece.